DNA grading of oral squamous carcinomasA preliminary report

DNA grading of oral squamous carcinomas

R. Chatelain, B. Hoffmeister 1, F. H~irle 1, A. B6cking and C. Mittermayer Department of Pathology, Aachen University of Technology, FRG; 1Department of Oral and Maxillofacial Surgery, University of Kiel, FRG

A preliminary report R. Chatelain, B. Hoffmeister, E Hdrle, A. B6cking and C. Mittermayer: D N A grading of oral squamous carcinomas - a preliminary report. Int. J. Oral Maxillofac. Surg. 1989; 18." 43-46. Abstract. The prognostic validity of the DNA-malignancy-grade ( D N A - M G ) , ranging on a continuous scale from 0.00 to 3.00 was tested in a preliminary study on 7 patients with oral squamous cell carcinomas. M o n o l a y e r smears were prepared after a cell separation procedure from paraffin-embedded surgical tumor specimens. Feulgen staining was performed automatically in a modified Shandon staining machine. A TV image analysis system with an automatic microscope (TAS plus, Leitz, F R G ) was used for D N A measurements. The D N A - M G revealed a strong correlation with the patients prognosis. 3 patients who died after having a tumor relapse (mean survival = 10 months) had D N A grades greater than 1.40, whereas patients who survived without having a t u m o r relapse (mean survival = 53 months) revealed D N A grades below 1.00. Additionally, the DNA-malignancygrade was closely correlated with the histopathologic malignancy grades.

Grading and staging of malignant tumors should provide relevant information for the clinical management and treatment planning of cancer patients. This also holds for oral carcinoma patients. During the past decades, various attempts have been made to establish histopathological grading systems for oral and laryngeal squamous cell carcinomas16, 20,23 26,28. However, grading systems based on the subjective evaluation of morphologic criteria alone are often not sufficiently reproducible. This h a s also been demonstrated for the JAKOBSSONet al? 4 grading system 22. Additionally the morphologic evaluation of oral cancer biopsy specimens tends to be not representative for the whole tumor due to the heterogenous cell appearance, with different degrees of differentiation within the tumor 2°. As for the prognostic validity, contradictory results have been published both for the BRODERS16 and for the JAKOBSSON et al. 24 grading systems ~-3,22, 23, 25, 26 Therefore, a grading system is needed, which is based on a non-morphologic parameter. The nuclear D N A content represents such a parameter, which is easy to quantitate and which has been shown to correlate with the malignant potential in various tumors 4, 15, and also in oral and laryngeal carci-

nomas 7, 19,21.27.29 The aim of this preliminary study was to test the prognostic validity of the previously developed DNA-malignancy-grade ( D N A - M G ) 6, ~0 for oral squamous cell carcinomas.

Material and Methods Surgical specimens of 7 male patients with oral squamous cell carcinomas (Table 1) were investigated, which were collected in the Department of Oral and Maxillofacial Surgery, University of Kiel, from 1980 to 1985. The mean age of the patients at diagnosis w a s 57.6 years. The tumor sites were: floor of the mouth (n - 3), anterior 2/3 of tongue (n = 2), buccal mucosa (n = 1) and maxillary alveolar ridge (n = 1). The follow-up period was up to 10 years. The TNM staging was performed according to the UICC 1987, resulting in 1 patient with T1 NO M0, 1 with T2 NO M0, 4 with T2 N1 M0 and 1 patient with a T3 NO M0 tumor (Table 1). The therapy comprised local excision in the T2 NO M0 and the T1 NO M0 patients, tumor excision and neck dissection in all T2 NI M0 patients, and resection of the maxillary bone in the T3 NO M0 patient. Hematoxylin and eosin stained sections of the paraffin-embedded material were reviewed in the Department of Pathology, University of Kiel, in order to obtain a histopathologic grading according to BRODERS1°. Monolayer smears were prepared from the

Key words: oral squamous cell carcinomas; DNA-imagecytometry; DNA-grading of malignancy. Accepted for publication 25 July 1988

paraffin-embedded material by mechanical and enzymatic cell separation procedures according to DELGADOet al, ~8. Feulgen staining w a s then performed automatically in a modified Shandon M24 staining machine as described in detail previouslyg, ~7. A TV image analysis system with an automatic microscope was used for interactive nuclear DNA measurements (MIAMEDDNA, Leitz) (5). Fig. 1 shows the new modular TV image analysis system for rapid interactive DNA measurements H. Measurements of at least 20 lymphocytes of the same smear served as an individual and tissue specific reference system for the determination of the diploid (2c) peak. After random measurements of 100 tumor cells detected morphologically on the TV screen, the computer loads an algorithm program for the computatation of the DNA-malignancy-grade (DNA-MG) 6, ~0 The DNA-MG is based on the variance of the single cell DNA contents around the 2c peak, the 2c deviation index (2cDI). The 2cDI ranges on a continuous scale from 0.00 to 51. A 2cDI of 51 was the highest value measured so far in an osteosarcoma 6. The DNA-MG finally represents a logarithmic transformation of the 2cDI: DNA-MG=3 x log (2cDI+ 1)/log 51. The resulting continous DNA-MG scale ranges from 0.00 to 3.00.

Results At the end of the follow-up, 4 patients were still alive and did not reveal tumor

44

Chatelain et al.

Table 1, Data file of 7 male patients with oral squamous cell carcinomas, including age, status (A = alive, D = dead), relapse of primary tumor, TNM stage, histologic malignancy grade according,to BRODERS~6,DNA-Malignancy-Grade (DNA-MG), site of primary tumor and therapy

Patient no.

Age (years)

1 2 3 4 5 6 7

80 59 41 51 61 71 40

Status A A A A D D D

Survival Tumor TNM (months) relapse stage 9l 85 45 80 8 8 14

No No No No Yes Yes Yes

relapse. 3 patients had died after a tumor recurrence (Table 1). The survival times after diagnosis ranged from 8 to 91 months (mean 47 months). The histopathologic grading resulted in 2 patients with grade I, 2 with grade II and 3 with grade III tumors (Table 1). The DNA-malignancy-grades (DNA-MG) ranged from 0.00 to 1.61 on the continuous scale from 0.00 to 3.00. The 4 patients with DNA-MG's < 1.00 (patients nos. 1 4 in Table 1) did not have a tumor relapse and were still alive at the end of the follow-up. Their survival times ranged from 45 to 91 months (mean = 75 months). Patient no. 1 with a DNA-MG of 0.00 revealed the longest survival time of 91 months. On the other hand, the 3 patients with DNA-MG's > 1.40 (nos, ~ 7 in Table 1) died within 8 to 14 months after having a tumor recurrence. A good correlation resulted between the D N A - M G and the histopathologic grading results. Grade I tumors corresponded to DNAMG's of 0.00 and 0.48, grade II to DNA-MG's of 0.67 and 0.98 and grade III tumors to DNA grades of 1.45, 1.46 and 1.61, respectively. All patients with NO stage were in the low D N A - M G range, 3 Nl-stage patients revealed high DNA grades, whereas 1 N1 patient had a low DNA-MG (0.67). This patient is still alive after 45 months without tumor relapse. Both tongue and floor of the mouth tumors were in the high D N A - M G range. The buccal mucosa, the maxillary alveolar ridge and floor-of-themouth tumors had l o w D N A grades.

T3NOM0 T1NOM0 T2NIM0 T2NOM0 T2N1M0 T2N1M0 T2NIM0

Histologic grade DNA-MG I I II II III III III

0.00 0.48 0.67 0.98 1.45 1.46 1.61

Tumor site

Therapy

maxillary alveolar ridge floor of mouth floor of mouth buccal mucosa floor of mouth tongue tongue

high representativity (grades obtained from'b[ops!es should be representative for the tumO~r as a whole); (3) high reproducibility (low intra- and interobserver variation). Histopathologic grading systems are based on the subjective evaluation of morphologic parameters and are therefore often not sufficiently reproducible. This also counts for the multifactorial grading system of JAKOBSSON et a/. 24, which has been modified by LUND et

maxillary bone resection local tumorexcision tumorexcision, neck dissection local tumorexcision tumorexcision, neck dissection tumorexcision, neck dissection tumorexcision, neck dissection

al.~,5, 26 for oral cancer grading purposes 22. As for the representativity of the grading results, FISHER2° indicated that with his modified JAI~OBSSONet a l ? 4 grading system, the malignancy grade of biopsy tissue tended to be lower than the grade of the surgical specimen. The BRODERS16 grading system has been used for many years in squamous cell carcinomas. However, the correlation between the BRODERS16 grades and the patient prognosis has been pro-

Discussion

The purpose of tumor staging and grading of tumor malignancy is to provide basic information for the clinical management of tumor patients. A malignancy grading system should tt~erefore be of: (1) high prognostic validity; (2)

Fig. 1. TV image analysis system (MIAMED-DNA, Leitz, FRG) with automated microscope

for rapid interactive DNA cytometry. Feulgen stained cell nuclei are selected with a mouse on the monitor for DNA measurements. All microscope functions are driven by monitor sensefield programs.

Oral squamous cell carcinomas

ven to be of limited significance3. Controversial results were reported for the Jakobsson grading, which has been modified by various authors. While some authors have reported a close correlation between patient prognosis and histologic grade 1, 2, 23, others could not confirm these results22,25,26 Thus, other grading systems based on non-morphologic parameters should be developed to allow more reproducible and prognostically valid grading of tumor malignancy. A non-morphologic parameter, which can reproducibly be determined and which is known to correlate well with the malignant potential of a tumor, is the nuclear DNA content4, 15.The prognostic significance of the nuclear DNA has already been demonstrated in squamous cell carcinomas of the oral cavity and the larynx19,2i, 27,29 However, instead of deriving a numerical prognostic index from the single cell DNA contents, these authors merely described the D N A distribution by means of histograms and grouped the cases subjectively according to different patterns. In order to create a DNA-based numerical grading system, we defined the DNA-malignancy-grade (DNA-MG), which is based on the variance of the tumor cell D N A contents around the diploid peak, the 2c deviation index (2cDI) (6, 10), The DNA-MG has been demonstrated to be of prognostic significance in malignant lymphomas8, 9, prostate carcinomas ~2, carcinomas of the breast 14 and also in squamous cell carcinomas of the larynx7. The reproducibility and representativity of the DNA-MG was high in various malignant tumors and was found to be superior to the compared morphologic gradingS, 13. Manual Feulgen staining and the application of single cell DNA cytometry with conventional microphotometers was until now too time consuming for routine purposes. For this reason, we developed a temperature-controlled staining machine for automated Feulgen staining (17), and use o f a TV image analysis system for rapid interactive DNA cytometry 5, 11. With this system, measurements of 100 tumor cells and reference cells including the computation and printout of the DNA-MG takes up to 40 min for one smear. This investigation represents a preliminary study on 7 patients with oral squamous cell carcinomas prior to a study on large series of patients of the

German-Austrian-SwissAssociation for head and neck tumors (DOSAK). The results revealed a strong correlation of the D N A - M G with the patients prognosis. Statistical tests, however, were not applied due to the low number of patients. 4 patients had DNA-MG's below 1.00. These patients survived (mean survival time 75 months) and did not have a tumor relapse. 3 patients with DNA-MG's above 1.40 died of their tumors (mean survival time 10 months) after tumor relapse. The histopathologic grading also correlated with the patient prognosis, and the histologic grades corresponded to the level of the DNA-Malignancy-Grade. No strict correlation was found between the TNM stage and the DNA-MG, as 3 T2 N1 M0 tumors revealed the highest DNA grades; however, in 1 T2 N1 M0 tumor, a low D N A - M G was found and the patient was one of the long survivors. The 3 NO patients indeed were also long survivors and revealed low DNA-malignancy-grades. In conclusion, it could be demonstrated in this preliminary study, that the DNA-malignancy-grade might provide clinically relevant prognostic information for the individual oral carcinoma patient, independent of the T N M staging. These results should indeed be investigated on a larger series of patients prior to a definite interpretation of the results. Acknowledgement - We want to thank Pro-

fessor Lennert, Institute of Pathology, University of Kiel, FRG, who made the histological diagnosis and grading and was so kind as to provide us with the praffin-embedded material for D N A measurements.

References

1. Anneroth, G., Batsakis, J. G. & Luna, M.: Malignancy grading in oral squamous cell carcinomas in the floor of the mouth related to clinical evaluation. Seand. J. Dent. Res. 1986: 94: 347-350. 2. Anneroth, G., Hansen, L. S. & Silverman, S.: Malignancy grading in oral squamous cell carcinoma (I). Squamous cell carcinoma of the tongue and floor of the mouth: histologic grading in the clinical evaluation. J. Oral. Pathol. 1986: 15: 162-168. 3. Arthur, K. & Farr, H. W.: Prognostic significance of histologic grade in epidermoid carcinoma of the mouth and pharynx. Am. J. Surg. 1972: 124: 489~492. 4. Atkin, N. B. & Kay, R.: Prognostic significance of modal DNA value and other factors in malignant tumors, based on

45

1465 cases. Br. J. Cancer 1979: 40: 210-221. 5. Auffermann, W., Repges, R. & B6cking, A.: Rapid diagnostic DNA cytometry with an automatic microscope and a TV image analysis system. Analyt. Quant. Cytol. 1984: 6: 179-188. 6. B6cking, A., Adler, C. R, Common, H. D., Hilgarth, M., Granzen, B. & Auffermann, W.: Algorithm for a DNA cytophotometric diagnosis and grading of malignancy. Analyt. Quant. CytoL 1984: 6: 1-7. 7. B6cking, A., Auffermann, W., Vogel, H., Schl6ndorff, G. & Goebbels, R.: Diagnosis and grading of malignancy in squamous epithelial lesions of the larynx with DNA cytophotometry. Cancer 1985: 56: 1600-1604. 8. B6cking, A., Chatelain, R., Auffermann, W., Krfiger, G. R. IV., Asmus, B., Wohltmann, D. & Schuster, C.: DNA grading of malignant lymphomas. I. Prognostic significance, reproducibility and comparison with other classifications. Anticancer Res. 1986: 6: 1205-1216. 9. B6cking, A., Chatelain, R., Auffermann, W., L6hr, G. W., Reif, M., Rossner, R. & Becker, H.: DNA grading of malignant lymphomas. II. Correlation with clinical parameters. Anticancer Res. 1986: 6: 1217-1224. 10. B6cking, A. & Auffermann, W.: Algorithm for DNA cytophotometric diagnosis and grading of malignancy. Analyt. Quant. Cytol. Histol. 1986: 6: 363. 11. B6cking, A., Sanchez, L., Stock, B., Milller, W.: Automated DNA cytophotometry. Lab. Practice 1987: 36:73 74. 12. B6cking, A., Chatelain, R., Orthen, U., Gien, G., yon Kalekreuth, G. & Wohltmann, J. D.: DNA grading of prostatic carcinoma prognostic validity and reproducibility. Anticancer Res. 1988: 8: 129-136. 13. B6cking, A., Chatelain, R., Homge, M., Daniel, R. & Wohltmann, D.: Representativity and reproducibility of the DNAMalignancy-Grade in different carcinomas. Analyt. Quant. Cytol. Histol. 1989: in press. 14. B6cking, A., Chatelain, R., Biesterfeld, S., Noll, E., Biesterfeld, D., Wohltmann, D. & Goecke, C.: DNA grading of breast cancer. Prognostic validity, reproducibility and comparison with other classifications. Analyt. Quant. Cytol. Histol. 1989: in press. 15. B6hm, N. & Sandritter, W.: DNA in human tumors: a cytophotometric study. Curr. Top. Pathol. 1975: 60: 151-219. 16. Broders, A, C.: Carcinoma of the mouth: types and degrees of malignancy. Ann. J. Roentgenol. Rad. Ther. Nucl. Med. 1927: 17: 90-93. 17. Chatelain, R., Willms, A., Auffermann, W. & B6cking, A.: Automated Feulgen staining with a temperature controlled staining machine. Analyt. Quant. Cytol. Histol. 1988: submitted.

46

Chatelain et al.

18. Delgado, R., Mikuz, G. & Hofstfidter, E: DNA - Feulgen cytophometric analysis of singel cells isolated from paraffin embedded tissue. Pathol. Res. Pract. 1984: 179: 92-94. 19. Feichter, G. E., Maier, H., Adler, D., Born, I. A., Abel, U., Haag, D. & Goerttler, K.: S - phase fractions and DNA ploidy of oropharyngeal squamous epithelium carcinomas compared with histologic grade, stage, response to chemotherapy and survival. Acta Otolaryngol. 1987: 104: 37~384. 20. Fisher, H. R.: Grading of biopsies of laryngeal carcinomas by multiple criteria. Can. J. Otolaryngol. 1975: 4: 881-884. 21. Goldsmith, M. M., Cresson, D. S., Postma, D. S., Askin, F. B. & Pillsbury, H. C.: Significance of ploidy in laryngeal cancer. Am. J. Surg. 1986: 152: 396-402. 22. Helweg-Larsen, K., Graem, N., Meistrup-Larsen, K. J. & Meistrup-Larsen, U.: Clinical relevance of histological gra-

23.

24.

25.

26.

ding of cancer of the larynx. Acta Pathol. MicrobioL Scand. 1978: 86: 499-504. Holm, L. E., Lundquist, P. G., Silversv/ird, C. & Sobin, A.: Histological grading of malignancy in squamous cell carcinoma of the oral tongue. Acta Otolaryngol 1982: 94: 185-192. Jakobsson, P. A., Eneroth, C. M., Killander, P., Moberger, G. & Martensson, B.: Histologic classification and grading of malignancy of the larynx. A pilot study. Acta Radiol. Ther. Phys. Biol. 1973: 12: 1-8. Lurid, C., S6gaard, H., Elbr6nd, O., J6rgensen, K. & Pedersen, A. P.: Epidermoid carcinoma of the lip. Histologic grading in the clinical evaluation. Acta Radiol. Ther. Phys. ~iol. 1975: 14: 465-474. Lund, C., S6gaard, H., Elbr6nd, O., J6rgensen, K. & Pedersen, A. P.: Epidermoid carcinoma of the tongue. Histologic grading in the clinical evaluation. Acta Radiol. Ther. Phys. Biol. 1975: 14: 513-521.

27. Nitze, H. R.: Cytological investigation of tumors with different clinical progressivity. Arch. Klin. Exp. Ohr. Nase. Kehl. Heilk. 1969: 193: 112-120. 28. Syrjfinen, K., Nuutintn, J. & K/irj~i, J.: Tumor differentiation and tumor host interactions as prognostic determinants in squamous cell carcinoma of the lip. Acta Otolaryngol 1986: 101: 152-160. 29. Tytor, M., Franzen, G. & Olofsson, J.: DNA pattern in oral cavity carcinomas in relation to clinical stage and histological grading. Pathol. Res. Pract. 1987: 182: 202-206. Address: C. Mittermayer Institut fiir Pathologie Klinikum der R W T H Pauwelsstrasse 5100 Aachen FRG